The rapid adoption of digital cameras and camcorders has resulted in explosive growth of personal photos and home video in digital form. There is a huge demand for new tools and systems to enable an average user to efficiently and effectively capture, process, manage, author and share digital media contents. Although video authoring tools exist, video authoring remains as a tedious and extremely time consuming task that often requires trained professional skills. Conventional content analysis techniques have tried to address these issues; however, these are difficult problems to resolve due to the difficulties in bridging the gap between high-level semantics with low-level features.